A-438079 – Oh Fmk Caspase Inhibitorvi

Astrocytic rather than neuronal P2X7 receptors modulate the function of the tri-synaptic network in the rodent hippocampus

Muhammad Tahir Khana, Jan Deussingb, Yong Tangc, Peter Illesa,c,⁎

A B S T R A C T

Whole-cell patch clamp recordings demonstrated that in the dentate gyrus (DG) as well as in the CA3 area of mouse hippocampal slices the prototypic P2X7 receptor (R) agonist dibenzoyl-ATP (Bz-ATP) induced inward current responses both in neurons and astrocytes. Whereas the selective P2X7R antagonist A438079 strongly inhibited both neuronal and astrocytic currents, a combination of ionotropic glutamate receptor (CNQX, AP-5) and GABAA-R (gabazine) antagonists depressed the Bz-ATP-induced current responses in the DG (granule cells) and CA3 neurons only. It was concluded that Bz-ATP activated astrocytic P2X7Rs and thereby released gluta- mate and GABA to stimulate nearby neurons. The residual A438079-resistant current response of astrocytes was suggested to be due to the stimulation of P2XRs of the non-P2X7-type. Further, we searched for presynaptic P2X7Rs at the axon terminals of DG and CA3 pyramidal neurons innervating CA3 and CA1 cells, respectively. Bz- ATP potentiated the frequency of spontaneous postsynaptic currents (sPSCs) in CA1 but not CA3 pyramidal cells. However, the Bz-ATP eﬀect in CA1 cells was inhibited by gabazine or the astrocytic toXin ﬂuorocitrate sug- gesting stimulation of P2X7Rs at stratum radiatum astrocytes located near to interneurons and synapsing onto CA1 neurons. Our data suggest that functional P2X7Rs are missing at neurons in the tri-synaptic network of the rodent hippocampus, but are present at nearby astrocytes indirectly regulating network activity.

Keywords:
ATP
Hippocampal neurons Hippocampal astrocytes Neuronal P2X7 receptors
Spontaneous postsynaptic currents

1. Introduction

The hippocampus has been shown to consist of three types of ex- citatory neurons, such as the granule cells of the dentate gyrus (DG), as well as the CA3 and CA1 pyramidal neurons, all of these synaptically interconnected with each other in a sequential manner (Fig. 1; Neves et al., 2008). The major sensory input of the hippocampus originates from neurons in layer II of the entorhinal cortex and terminates at the ﬁrst hippocampal relay station of DG granule cells. These cells project via their axons termed mossy ﬁbers to CA3 pyramidal neurons, which in turn innervate the ipsilateral CA1 pyramidal cells through Schaﬀer collaterals, and the contralateral CA1 pyramidal neurons through commissural ﬁbers. There exist also associative networks inter- connecting CA3 cells on the ipsi-and contralateral sides. Based on the excitatory tri-synaptic network and the intertwined inhibitory interneurons, the hippocampus is ideally suitable to store and replay episodic, spatial and contextual information through long-range and local circuit interactions (Ekstrom and Ranganath, 2017; Voss et al., 2017).
Intracellular ATP is used for the storage and subsequently the lib- eration of energy all over the animal kingdom. However, it may also pass the cell membrane by diverse mechanisms, accumulate in the ex- tracellular space and thereafter subserve intercellular communication (Burnstock and Verkhratsky, 2009). After exocytotic, vesicular release from neuronal terminals ATP functions as a transmitter or co-trans- mitter in the peripheral and central nervous systems (Burnstock, 2006). ATP-sensitive P2Rs have been classiﬁed into two groups: the ligand- gated P2X-type and the G-protein-coupled P2Y-type (Abbracchio and Burnstock, 1994; Fredholm et al., 1994). P2XRs mediate rapid, while P2YRs mediate slow signaling (Illes and Alexandre Ribeiro, 2004).
Within the P2X-type, 7 subtypes have been identiﬁed (P2X1–7), which form homotrimeric or heterotrimeric assemblies and mediate rapid transmembrane ﬂuXes of cations. The P2X7R is unique in occurring only as a homotrimer and causing, in addition to immediate eﬀects, also molecular changes on a much longer time-scale, such as proliferation and necrosis/apoptosis (Surprenant et al., 1996; North, 2002).
There is an ongoing discussion whether P2X7Rs are located in the CNS only on immunocompetent cells (microglia, astrocytes) (Volonte et al., 2012; Bhattacharya and Biber, 2016; Illes et al., 2012, 2017) or also on neurons (Miras-Portugal et al., 2016, 2017). Indispensable ex- perimental tools for investigating P2X7R function are mice in which this receptor has been genetically deleted. Originally, two strains of P2X7 KO mice were available: the Glaxo line generated by Chessell et al. (2005) and the Pﬁzer line generated by Solle et al. (2001). However, it was reported later that in both knock-out strains several still functional splice variants escape deletion thereby rendering these mice unsuitable to decide whether neuronal P2X7Rs exist (Nicke et al., 2009; Masin et al., 2012). Much stronger evidence for the presence of P2X7Rs in hippocampal CA1 cells and DG granule neurons was supplied by using a transgenic P2RX7-EGFP (enhanced green ﬂuorescence pro- tein) mouse (Jimenez-Pacheco et al., 2016; Miras-Portugal et al., 2017). Further, of great signiﬁcance, in this respect, was a recently generated knock-in mouse in which a humanized P2RX7 allele is accessible to spatially and temporally controlled Cre recombinase-mediated in- activation (Metzger et al., 2017a). By selective disruption and assess- ment of human P2RX7 expression in diﬀerent brain regions, it was possible to demonstrate that the P2X7R-mRNA is speciﬁcally expressed in glutamatergic CA3 neurons of the mouse hippocampus.
We have previously shown that dibenzoyl-ATP (Bz-ATP), with preference for P2X7Rs, caused inward current in CA1 neurons of rat hippocampal slices; this eﬀect fully depended on the release of gluta- mate and GABA (Ficker et al., 2014). It was concluded that P2X7Rs are situated at astrocytes which release amino acid transmitters to stimu- late or inhibit the neighboring pyramidal cells. Now we investigated in wild-type mice, whether DG granule cells and CA3 pyramidal cells exhibit a similar response to Bz-ATP as CA1 cells do. Moreover, we studied whether Bz-ATP causes an increase of the spontaneous release of glutamate and GABA from nerve terminals innervating CA3 and CA1 neurons by mossy ﬁbers and Schaﬀer collaterals, respectively. It ap- pears that neuronal P2X7Rs fail to cause electrophysiologically mea- surable eﬀects under the conditions of the present experiments.

2. Materials and methods

2.1. Preparation of hippocampal brain slices

Hippocampal brain slices were obtained from C57BL/6J mice (own breed). All animal use procedures were approved by the relevant Committee of Animal Protection. Mice of either sex (12–16 days old) were decapitated under CO2 anesthesia, and their hippocampal slices were prepared and then used immediately for patch-clamp recordings as recently described (Oliveira et al., 2011; Ficker et al., 2014). In short, after decapitation, the brain was placed into ice-cold, oXygenated (95% O2 + 5% CO2) artiﬁcial cerebrospinal ﬂuid (aCSF) of the following composition (in mM): NaCl 126, KCl 2.5, CaCl2 2.4, MgCl2 1.3, NaH2PO4 1.2, NaHCO3 25, and glucose 11; pH 7.4 adjusted with NaOH. Transverse slices of a nominal thickness of 200 μm were cut from the hippocampus using a vibratome (VT1200, Leica Biosystems).

2.2. Whole-cell patch-clamp recordings and drug application

Hippocampal slices were superfused at a speed of 3 ml/min in an organ bath with 95% O2 plus 5% CO2-saturated aCSF at room tem- perature. To create low X2+ solution, MgCl2 was omitted from the medium and CaCl2 concentration was decreased to 0.5 mM. Neurons and astrocytes in the DG, CA3 and CA1 areas were visualized with an upright interference contrast microscope and a 40× water immersion objective (AXioskope FS, Carl Zeiss). Patch pipettes were ﬁlled with intracellular solution of the following composition (in mM): K-gluconic acid 140, NaCl 10, MgCl2 1, HEPES 10, EGTA 11, Mg-ATP 1.5, Li-GTP 0.3; pH 7.2 adjusted with KOH. Pipettes (4–7 MΩ resistances) were pulled by a horizontal micropipette puller (P-97, Sutter Instruments) from borosilicate capillaries. The resting membrane potential of the cells was checked in the current-clamp mode of the patch-clamp am- pliﬁer (Multiclamp 700A, Molecular Devices) immediately after estab- lishing whole cell access.
Astrocytes were discriminated from neurons by their failure to ﬁre action potentials. For this purpose hyper- and depolarizing current pulses (−80, −20, 40, 100, 160 pA) were injected into the respective cells. In some of the experiments, we calculated the input resistance of CA1 and CA3 pyramidal neurons as well as of DG granule cells by in- jecting a hyperpolarizing current pulse of −80 pA amplitude. The resting membrane potentials (Vm) of the three cell types did not diﬀer from each other (CA1, 65.7 ± 2.8 mV; CA3, 63.0 ± 1.7 mV; DG, 64.3 ± 4.0 mV; n = 15 each; P > 0.05). By contrast, the input re- sistance (Rm) of the DG cells (620.8 ± 64.0 MΩ) was higher than that of the CA1 (367.0 ± 47.9 MΩ; P < 0.05) but not that of the CA3 (442.8 ± 70.5 MΩ; P > 0.05) pyramidal cells (n = 15 each). Astrocytes had higher Vm and much lower, although divergent Rm va- lues, because they belonged in variable numbers to the outwardly rectifying, variably rectifying and passive types of glia (Oliveira et al., 2011, Suppl. Table 1).
Then, in the voltage-clamp recording mode of the ampliﬁer, the holding potential of neurons and astrocytes was set at −70 mV and −80 mV, respectively. Multiclamp and pClamp 10.4 software (Molecular Devices) were used to store the recorded data to perform online analysis/ﬁltering and to trigger the application system used.
Spontaneous postsynaptic currents (sPSCs) were recorded from neurons at the holding potential of −70 mV. The sPSCs were analyzed by means of the pClamp 10.4 software package, by detecting ampli- tudes exceeding the detection threshold set at three times the standard deviation above the baseline noise of the recordings. False positive noise-triggered ﬂuctuations and signals with a non-monotonic rising phase and/or additional events within the decay phase were rejected on visual inspection (< 2%). Since sPSCs compose of both action potential- induced and spontaneous vesicular glutamate/GABA release, the blockade of GABAA receptors by gabazine left us with pure glutama- tergic currents. 2.3. Drug application and experimental protocols All drugs were pressure ejected locally by means of a computer- controlled DAD-12 superfusion system (ALA Scientiﬁc Instruments). The drug application tip touched the surface of the brain slice and was placed within 100–150 μm from the patched cell. When current responses to various drugs (Bz-ATP, AMPA, NMDA, muscimol) were determined, selected agonist concentrations were su- perfused for 10 s, every 2 min. In another series of experiments, Bz-ATP was pressure ejected siX-times in total, twice before, twice during, and twice after the application of A438079 or a combination of CNQX, AP- 5, and gabazine for 4 min. The changes in the Bz-ATP current amplitude were calculated as a percentage of the mean of the two pre-antagonist (control) current amplitudes. Time-independent stability of control re- sponses was shown by conﬁrming recovery after washout of the an- tagonists. sPSCs were recorded for 5 min in the absence and for another 5 min in the presence of Bz-ATP. The mean of the pre-drug (control) sPSC amplitude and frequency was considered as 100% and was used as a reference value for calculating the percentage change caused by Bz- ATP. Once again, reversibility of the Bz-ATP eﬀect was demonstrated by recording sPSCs for another 5 min after washing out Bz-ATP. Two sets of hippocampal slices were prepared and one of them was incubated in ﬂuorocitrate-free low X2+ aCSF for 2 h, while the other one was incubated in ﬂuorocitrate (100 μM)-containing aCSF for the same period of time. Then, the eﬀects of Bz-ATP on the sPSC amplitude and frequency in hippocampal slices treated or untreated with ﬂuor- ocitrate were investigated. 2.4. Materials The following drugs were used: DL-ﬂuorocitric acid barium salt (Sigma- Aldrich); (S)-α-amino-3-hydroXy-5-methyl-4-isoXazolepropionic acid (S- AMPA), 2′(3′)-O-(4-benzoylbenzoyl)adenosine-5′-triphosphate tri(triethy- lammonium) salt (Bz-ATP), 3-[[5-(2,3-dichlorophenyl)-1H-tetrazol-1-yl] methyl]pyridine hydrochloride (A438079), 6-cyano-7-nitroquinoXaline- 2,3-dione (CNQX), D-(-)-2-amino-5-phosphonopentanoic acid (AP-5), ga- bazine, muscimol, N-methyl-D-aspartic acid (NMDA) (Tocris Bioscience). 2.5. Statistics Concentration-reponse curves were ﬁtted with the SigmaPlot 13.0 (SPSS) software using the three parametric Hill equation where I represents the peak current elicited by a certain concentration of Bz-ATP [A], Imax is the peak current evoked by the maximal eﬀective concentration of Bz-ATP, EC50 is the half maximal eﬀective con- centration of Bz-ATP, and nH is the Hill coeﬃcient. Means ± S.E.M. are given throughout. SigmaPlot 13.0 was also used for statistical evaluation. We tested for and found that, when using parametric tests, all sampled distributions satisﬁed the normality and equal variances criteria. Multiple comparisons between data were performed by one way analysis of variance (ANOVA) followed by the post hoc Holm-Sidak test. Two data sets were compared by using the Student’s t-test or the Mann-Whitney Rank-Sum test as appropriate. A probability level of 0.05 or less was considered to be statistically sig- niﬁcant. 3. Results 3.1. Agonist-induced currents in CA3 neurons and astrocytes In a ﬁrst series of experiments, we established concentration-re- sponse relationships for Bz-ATP (100–3000 μM) on CA3 neurons and astrocytes in a low Ca2+/no Mg2+ (low X2+) aCSF (Fig. 2A, B). The decrease of divalent cations in the external medium is known to in- crease the sensitivity of P2X7Rs to its agonists (Virginio et al., 1997) probably by reliving an allosteric block exerted on the receptor by Ca2+ and Mg2+ (Coddou et al., 2011). Therefore, most experiments in- vestigating the eﬀects of Bz-ATP on neuronal and astrocytic currents in CA3 and DG areas were made under low X2+ conditions. The inspection of Fig. 2B shows that Bz-ATP causes at a concentration of 3000 μM larger current responses at neurons than at astrocytes (−182.5 ± 40 pA vs. −95.7 ± 14.0 pA, n = 10 each, P < 0.001). Comparison of the Imax values (neuron, −181.0 ± 20.8 pA; astrocyte, 95.7 ± 0.1 pA, P < 0.001) led to the same conclusion. By contrast the EC50 values (neuron, 587.9 ± 165.3 μM; astrocyte, 508.4 ± 1.6 μM, P > 0.05) did not diﬀer from each other in a statistically signiﬁcant manner.
Then, we determined the eﬀects of the excitatory amino acid ago- nists AMPA and NMDA, and the inhibitory amino acid agonist mus- cimol at identical concentrations of 100 μM on neurons and astrocytes (Fig. 2C, D). It was found that AMPA stimulated both neurons and as- trocytes, although neurons with a much higher potency; NMDA acted only at neurons, and muscimol had almost identical eﬀects at the two cell types (Fig. 2D).
Our next aim was to conﬁrm that Bz-ATP activates P2X7Rs in CA3 neurons and astrocytes. In fact, the highly selective P2X7R antagonist A438079 (10 μM) considerably inhibited the eﬀects of Bz-ATP (1000 μM) both on neurons and astrocytes (Fig. 3A, C, D). A combination of antagonists for AMPA- (CNQX; 10 μM), NMDA- (AP-5; 50 μM) and GABAA-Rs (gabazine; 10 μM) depressed the Bz-ATP-induced current in neurons to an extent similar to that caused by A438079 (Fig. 3B, C). At the same time this antagonist cocktail did not alter the eﬀect of Bz-ATP on astrocytes (Fig. 3D). Therefore, we concluded that P2X7R activation most likely releases astrocytic glutamate and GABA onto CA3 neurons.

3.2. Eﬀect of Bz-ATP on sPSCs in CA1 neurons and stratum radiatum interneurons

CA3 and CA1 neurons are interconnected by the Schaﬀer collateral axon trajectories. Hence, stimulation of the glutamatergic CA3 neurons is expected to result in CA1 neurons in an increase of the frequency but not the amplitude of spontaneous excitatory postsynaptic currents (sEPSCs). On the other hand, Schaﬀer collaterals also terminate on GABAergic interneurons located below the CA1 band in the stratum radiatum. A stimulation of CA3 neurons thereby activates GABAergic (non-pyramidal) interneurons as well, which in turn project to CA1 pyramidal neurons and cause an increase of the frequency of sponta- neous inhibitory postsynaptic currents (sIPSCs). In a ﬁrst approach, we measured the pattern of spontaneous inward currents without diﬀer- entiating pharmacologically (by the use of selective antagonists for excitatory or inhibitory amino acid transmitters) or biophysically (by setting the holding potential to the equilibrium potential of one of these transmitters).
statistically signiﬁcant diﬀerence from the eﬀect of Bz-ATP (3000 μM) in neurons. (C, D) The application of AMPA, NMDA and muscimol (100 μM each) caused inward currents. Representative recordings for neurons (C, upper panel) and astrocytes (C, lower panel). Mean ± S.E.M. current amplitudes in neurons (D, left panel) and astrocytes (D, right panel). The number of experiments is indicated in the respective columns. *P < 0.05; statistically signiﬁcant diﬀerence from the eﬀects of the two residual agonists. §P < 0.05; statistically signiﬁcant diﬀerence from the eﬀects of the respective agonists in neurons. Bz-ATP (300 μM) increased the frequency, but did not alter the amplitude of sPSCs (Fig. 4A–C). Fig. 4A shows excerpts of sPSC re- cordings before, during and after the application of Bz-ATP. Fig. 4B shows the average of all sPSCs recorded during two 5-min periods (before and after Bz-ATP application) for comparison. Finally, the amplitude (in pA) and the frequency (in events/min) of sPSCs were plotted for quantitative evaluation and for the documentation of the reversibility of Bz-ATP eﬀects. The prototypic P2X7R agonist Bz-ATP (300 μM) alone, or in the presence of, the A1 adenosine receptor antagonist DPCPX (0.1 μM), or the GABAA-R antagonist gabazine (10 μM) did not modify thesPSC amplitudes in a normal aCSF bath medium (Fig. 4D). Similarly, Bz-ATP (300 μM) failed to change the amplitude of sPSCs either alone, or when applied in the presence of gabazine (10 μM), A438079 (10 μM) or after a 2-h pre-incubation with the selective astrocytic toXin ﬂuorocitrate (10 μM; Clarke, 1991); all these experiments were made in a low X2+ su- perfusion medium. The sPSC frequency remained also stable in spite of applying Bz-ATP in the presence of DPCPX or gabazine, when super- fused with a normal Ca2+/Mg2+-containing external medium (Fig. 4E). However, in a low Ca2+/no Mg2+ containing extracellular medium, Bz- ATP increased the frequency of sPSCs. It was interesting to note that gabazine abolished this facilitatory eﬀect, indicating a selective inﬂu- ence of Bz-ATP on the inhibitory synaptic input manifested as sIPSCs. A438079 also abolished the facilitation of sPSC frequency proving the involvement of P2X7Rs. Because ﬂuorocitrate interfered with Bz-ATP, it appeared likely that an astrocytic signaling molecule mediated the Bz- ATP-induced frequency increase. This signaling molecule could stimu- late stratum radiatum interneurons and potentiate the resting vesicular release of GABA. Another possibility is that the astrocytes themselves release pre-formed packages of GABA onto CA1 pyramidal neurons. This second alternative is strengthened by our experiments showing that in a low X2+ external medium, Bz-ATP (300 μM) did not alter either the amplitude (before, −14.3 ± 1.4 pA; after, -12.6 ± 1.1 pA, n = 10, P > 0.05) or the frequency (7.7 ± 0.9 events/min; after, 8.7 ± 0.8 events/min, n = 10, P > 0.05) of sPSCs recorded in stratum radiatum interneurons. Thus, it can be concluded with reasonable certainty that in CA1 neurons the Bz-ATP-induced release of the glio- transmitter GABA was responsible for the sPSC frequency increase.

3.3. Bz-ATP-induced currents in DG neurons and astrocytes

Concentration-response curves for the eﬀect of Bz-ATP (100-3000 μM) in low X2+ aCSF at neurons and astrocytes of the DG showed that at 3000 μM Bz-ATP caused only −63.7 ± 10.0 pA (n = 11) current in neurons, but a much larger current in astrocytes (−253.4 ± 24.8 pA, n = 10; P < 0.01) (Fig. 5A, B). Comparison of the Imax values (neuron, −64.7 ± 2.6 pA; astrocyte, 254.2 ± 1.8 pA, P < 0.001) led to the same conclusion. By contrast the calculated EC50 values (neuron, 904.6 ± 83.1 μM; astrocyte, 1083.8 ± 11.7 μM, P > 0.05) did not diﬀer from each other in a statistically signiﬁcant manner.
The highly selective P2X7R antagonist A438079 (10 μM) con- siderably inhibited the eﬀects of Bz-ATP (1000 μM) both on neurons and astrocytes (Fig. 5C, D). A combination of antagonists for AMPA- (CNQX; 10 μM), NMDA- (AP-5; 50 μM) and GABAA-Rs (gabazine; 10μM) depressed the Bz-ATP-induced current in neurons, but did not alter it in astrocytes (Fig. 5C, D). These ﬁndings were similar to those made in CA3 neurons/astrocytes and allowed us to conclude that P2X7R ac- tivation at astrocytes most likely releases glutamate and GABA onto CA3 neurons.

3.4. Eﬀect of Bz-ATP on sPSCs in CA3 neurons

DG neurons project to CA3 neurons via relatively sparse mossy ﬁber trajectories. Bz-ATP (300 μM) did not change either the amplitude or the frequency of the sPSCs, in low X2+ aCSF as shown by a number of synaptic currents recorded at a slower time-scale, as well as by super- imposed synaptic currents recorded at a faster time-scale (Fig. 6A, B). The failure of Bz-ATP to act on sPSC amplitude and frequency was conﬁrmed by plotting mean ± S.E.M. data obtained from 6 cells (Fig. 6C).
Calculation of percentage changes of the sPSC amplitude and fre- quency showed that the failure of Bz-ATP in a low Ca2+/no Mg2+ ex- ternal medium to exert a modulatory eﬀect persisted in a normal ex- ternal medium as well (Fig. 6D, E). Further, the application of DPCPX (0.1 μM) or gabazine (10 μM) to the Bz-ATP containing external medium, did not uncover a previously absent P2X7R-mediated action.

4. Discussion

The main ﬁnding of this study is that in three hippocampal cell types (CA3 and CA1 neurons, DG granule cells) of wild-type mice we did not ﬁnd any indication for direct, P2X7R-mediated eﬀects. By contrast, we suggest that the neuronal changes are due to the occupa- tion of P2X7Rs at astrocytes which thereafter release glutamate and/or GABA onto neighboring neurons exerting indirect stimulation and in- hibition, respectively. It should be stated, however, that our data were collected on 12-16-days (P12-16) old mice and therefore caution is recommended when extending these ﬁndings to older mice or espe- cially humans. The postnatal development of laboratory rodents and humans occurs at diﬀerent pace; the rodent hippocampus and speciﬁ- cally the dentate gyrus proliferates maximally around P8, is still de- veloping at P20 and reaches mature-like structures in the ﬁrst to second postnatal month (Avishai-Eliner et al., 2002). The same stage of ma- turation is achieved in humans only a number of years after birth.

4.1. P2X7Rs at hippocampal neurons

There are relatively limited data available on the existence of P2X- type receptors on rat hippocampal neurons. It was reported that the stimulation of Schaﬀer collateral/commissural ﬁbers evoked in CA1 pyramidal cells EPSCs, the amplitude of which could be reduced by CNQX and AP-5 (Pankratov et al., 1998). However, these glutamatergic antagonists still left a small residual EPSC uninhibited. This synaptic response, just as the eﬀect of exogenous ATP was blocked by the non- selective P2R antagonist PPADS and was potentiated by the trace ele- ment Zn2+. Although the P2X receptor-subtype could not be de- termined, both the marked sensitivity to PPADS and the potentiation by Zn2+ appear to exclude the involvement of P2X7R s(Jarvis and Khakh, 2009; Huidobro-Toro et al., 2008).
As mentioned previously, CA3 pyramidal cell axons innervate both CA1 neurons and neighboring stratum radiatum interneurons (Freund and Buzsáki, 1996). It was shown that ATP and its enzymatically stable analogue ATP-γ-S facilitate the sEPSC frequency at interneurons, but not at CA1 neurons (Khakh, 2009; Khakh et al., 2003). The P2XR-type was assigned to the P2X2 class, because P2X2 null mice did not exhibit an sEPSC frequency increase in their interneurons after ATP stimula- tion. It was especially interesting that P2X2Rs were selectively ex- pressed only on axons projecting to one cellular target, but not on those projecting to another target.

4.2. Memory storage in hippocampal circuits

The cellular correlate of memory storage is thought to involve changes in synaptic strength manifesting in long-term potentiation (LTP) or long-term depression (LTD) (Nicoll, 2017; Lomo, 2018). Classic, N-methyl-D-aspartate (NMDA) receptor (R)-dependent LTP re- quires postsynaptic depolarization coupled with synaptic stimulation. This type of LTP can develop at the Schaﬀer collateral-CA1 synapse, whereas a presynaptic, non-NMDA-R-dependent LTP was observed be- tween mossy ﬁbers and CA3 neurons. Blockade of the above described, unidentiﬁed P2XR-type located at CA1 neurons dramatically po- tentiated the LTP determined in the presence of PPADS (Pankratov et al., 2002).

4.3. Astrocyte-neuron cross-talk in the hippocampus

4.3.1. Exocytotic gliotransmitter release

The present results challenge the assumption that P2X7Rs are lo- cated at the investigated neurons themselves; a more likely mode of action is the stimulation of astrocytic P2X7Rs which initiate the se- cretion of signaling molecules/transmitters acting on the nearby neu- rons. The exocytosis of transmitters, in particular that of ATP, has been shown to be triggered by both ionotropic and metabotropic Ca2+ sig- naling induced for instance by the stimulation of astroglial P2X7- (Pankratov and Lalo, 2014), NMDA- (Lalo et al., 2014) or cannabinoid CB1-Rs (Rasooli-Nejad et al., 2014).
The exocytotic secretion from astrocytes involves divergent secre- tory organelles (synaptic-like microvesicles, dense-core vesicles, lyso- somes, and exosomes; Bowser and Khakh, 2007). It was demonstrated 20 years ago that the soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor (SNARE) is present in cultured astrocytes (Parpura et al., 1995). Subsequently it was reported that astrocytes express proteins characteristic for neuronal synaptic vesicles such as vesicle-associated membrane protein-2 (VAMP2) and proteins that are found in exocytotic traﬃcking vesicles of non-neuronal cells such as secretory carrier membrane protein (SCAMP) and VAMP3. Cleavage of SNARE proteins with tetanus toXin or botulinum neurotoXins reduced glutamate and ATP release in astrocytes (Coco et al., 2003).
The hypothesis on the vesicular release of transmitters from astro- cytes has been repeatedly challenged by claiming that experimental tools in glia research only incompletely diﬀerentiate between astrocytes and neurons (Nedergaard and Verkhratsky, 2012). For example it has been shown that in the dnSNARE mouse model, cortical neurons also express this transgene in spite of the published assumption of a selective astrocytic localization (Fujita et al., 2014). However, there is a con- tinuous strong belief in and further support for the physiological im- portance of exocytosis of gliotransmitters, in particular ATP, in com- munication between astrocytes and neurons and modulation of synaptic eﬃcacy (Lalo et al., 2014).

4.3.2. Connexin hemichannels and pannexin channels as routes of gliotransmitter release

GABA (in contrast to glutamate, ATP and D-serine) has not been identiﬁed hitherto as an exocytotically released gliotransmitter (Zhang and Haydon, 2005), although it may leave astrocytes via e.g. a calcium- activated chloride channel, the bestrophin 1 channel (Jo et al., 2014; Woo et al., 2018). Hence, other ways of glutamate and ATP release from astrocytes have also to be considered. A non-vesicular pathway for ATP outﬂow from astrocytes appears to utilize hemichannels formed by the mammalian gap junction protein connexin-43 (CX43); another protein, pannexin-1 (Panx1), functions as a transmembrane channel without subserving intercellular communication (Giaume et al., 2013; Illes and Verkhratsky, 2016). CX43 hemichannels have been implicated in ATP release as inferred from the inhibition by certain gap junction channel blockers and connexin-mimetic peptides as well as in astrocytic cultures prepared from CX43 deﬁcient mice (Kang et al., 2008; Leybaert et al., 2003).
The long-term activation of P2X7Rs with high concentrations of ATP leads to the opening of a large membrane pore (Surprenant et al., 1996; Yan et al., 2010). It was suggested that pore opening is not an inherent property of the ATP-gated cationic channel but is due to recruitment of an accessory protein, the Panx1 channel (Pelegrin and Surprenant, 2006). It has to be noted that signiﬁcant doubts were raised against this hypothesis, because hemichannel antagonists, interference RNA targeting of Panx1, and colchicine treatment did not aﬀect P2X7R currents, although they inhibited the accompanying dye uptake into P2X7R expressing macrophages (Marques-da-Silva et al., 2011; Alberto et al., 2013). Nonetheless, ATP release has been shown to occur via both Panx1 channels (Dahl, 2015) and P2X7Rs themselves (Suadicani et al., 2006, 2009).
In spite of the overwhelming evidence for the participation of CX hemichannels and Panx channels in the astrocytic release of ATP, they are also instrumental in the outﬂow of glutamate into the extracellular space (Cheung et al., 2014; Montero and Orellana, 2015). With respect to the suggested coupling of P2X7Rs to Panx1 it is especially interesting to mention that the activation of this receptor in astroglia allows glu- tamate to transit the cell membrane (Duan et al., 2003; Fellin et al., 2006). In fact, Bz-ATP has been reported to trigger 2 diﬀerent gluta- mate mediated responses in CA1 pyramidal neurons in the rat hippo- campus. These were (1) transient inward currents due to calcium-de- pendent but P2X7R-independent exocytotic glutamate release from astroglial cells and (2) a sustained tonic current also due to glutamate release but dependent on P2X7R activation (Fellin et al., 2006).

4.3.3. P2X7R-mediated release of gliotransmitters

The concept of an astrocyte-neuron cross-talk via P2X7R-mediated release of glutamate and GABA is strengthened by previous experiments in the rat hippocampus and spinal cord dorsal horn (Ficker et al., 2014). Whereas the ATP- and Bz-ATP-induced current responses of CA1 neu- rons were abolished by the P2X7R antagonist A438079, as well as by the combination of ionotropic glutamate and GABAAR antagonists (AP- 5, CNQX, gabazine), P2X7R-mediated currents at astrocytes in the substantia gelatinosa of the rat spinal cord were inhibited only by A438079, but not by the above mentioned antagonist cocktail. A similar experimental schedule in the hippocampal dentate gyrus and CA3 areas of mice led to comparable results, although the inhibitory eﬀects of the pharmacological ligands used were less pronounced than in the rat hippocampus (Ficker et al., 2014; present results). Nonetheless, we concluded that astrocytic P2X7R activation stimulates granule cells and CA3 neurons via glutamate and GABA release from the neighboring glial cells. The lack of a complete inhibition by the respective antago- nists of ATP/Bz-ATP eﬀects is due to the fact that these agonists are non-selective for P2X7Rs and continue to activate the residual P2XR- types existing at the astrocytes under investigation (Illes et al., 2012: Boue-Grabot and Pankratov, 2017).
Notwithstanding the early identiﬁcation of P2X7Rs at microglial cells, the resident monocytes/macrophages of the CNS, strong evidence has accumulated on the location of this receptor type on neuroglial cells as well (Illes et al., 2012; Franke and Illes, 2014). Especially the re- cording of P2X7R-mediated currents in rodent cortical astrocytes (Duan et al., 2003; Norenberg et al., 2010) and in astrocytes patch-clamped in acute brain slices of rats (Oliveira et al., 2011) supplied convincing arguments for the functionality of this receptor-type. The P2X7R identity of these currents is supported by the following data: (1) Bz-ATP was about 10-times more potent than ATP itself; (2) removal of divalent cations from the extracellular medium greatly potentiated the responses to ATP; (3) selective P2X7R antagonists inhibited the Bz-ATP-induced currents; and (4) high concentrations of ATP failed to induce currents in a classic P2X7R knockout mice (Illes et al., 2012).
DG granule cells innervate with their axons CA3 neurons which in turn innervate CA1 neurons both directly and indirectly via GABAergic interneurons of the stratum radiatum. Therefore, we measured in these target neurons the changes in sPSC frequency but not amplitude, sup- posed to be a measure of the spontaneous release of excitatory/in- hibitory transmitters. Bz-ATP indeed failed to increase the sPSC fre- quency in CA3 neurons. In CA1 pyramidal cells there was a frequency increase which, however, appeared to be due to the stimulation of as- trocytic rather than neuronal P2X7Rs. Further, it depended on the re- lease of GABA rather than that of glutamate and was blocked by the selective astrocytic toXin ﬂuorocitric acid. Hence the existence of pre- synaptic P2X7Rs situated at the axon terminals of granule cells or CA3 neurons appeared to be highly unlikely. Although previously Bz-ATP has been shown to depress ﬁeld EPSPs at mossy ﬁber-CA3 pyramidal cell synapses, and was suggested to act at presynaptic P2X7Rs (Armstrong et al., 2002), this inhibition was found later to be mediated by A1 type adenosine receptors (Kukley et al., 2004). Bz-ATP was ap- parently enzymatically degraded to Bz-adenosine which displaced en- dogenous adenosine from its intracellular pools.
In addition to astrocytes, a signiﬁcant amount of extracellular ATP can be released from microglia, in particular during neuroinﬂammation (Beamer et al., 2016; Boue-Grabot and Pankratov, 2017). Even an exocytotic release of ATP has been shown to occur from cultured mi- croglial cells after ionomycin-induced increase of intracellular Ca2+ (Imura et al., 2013). However, electrophysiological work demonstrated that the preferential partners of neurons in brain slice preparations, representing near physiological conditions, are astrocytes rather than microglial cells (Oliveira et al., 2011; Ficker et al., 2014). Cells en- dowed with functional P2X7Rs were labelled by Lucifer Yellow (LY)- diﬀusion from the recording pipettes; the microglial marker GSA-B4 never stained the LY-labeled astroglial cells, although GSA-B4-im- munopositive microglia were present in the investigated brain slices. Microglial cells may exhibit morphological characteristics somewhat similar to astroglia after LY ﬁlling and also fail to ﬁre action potentials in response to depolarizing current injection (Brockhaus et al., 1993; Boucsein et al., 2000). However, in voltage-clamp recordings, resting microglia express a typical inwardly rectifying current pattern and ac- quire an additional outwardly rectifying current component only after cell activation, for example, following facial nerve axotomy (Boucsein et al., 2000). Thus, immunohistochemistry in conjunction with voltage- clamp recordings allowed the distinction of astroglial cells from mi- croglia.

4.4. Doubts concerning the reliability of recently generated transgenic and knock-in-animals to prove the existence of neuronal P2X7Rs

As mentioned in the Introduction, recent experiments with a transgenic P2RX7-EGFP mouse led to the conclusion that hippocampal CA1 cells and DG granule neurons are endowed with P2X7Rs (Jimenez- Pacheco et al., 2016; Miras-Portugal et al., 2017). However, the strong expression at the mRNA level in the CA3 area in a knock-in-mouse (Metzger et al., 2017a, 2017b) is not reﬂected by EGFP expression in this reporter mouse line. Further, data concerning the neuronal locali- zation of P2X7Rs generated by means of these two mouse lines could not be conﬁrmed by our present electrophysiological experiments car- ried out in the wild type animals.

5. Conclusions

Based on our experiments we propose that hippocampal neurons do not possess functional P2X7Rs. By performing measurements under low X2+ conditions which are supposed to amplify P2X7R-mediated eﬀects, this conclusion obtains further support. However, under (quasi)phy- siological conditions P2X7Rs at hippocampal astrocytes are the targets of endogenously released ATP; after the activation of these receptors, astrocytic signaling molecules are released which might stimulate nearby neurons in order to subserve a pertinent cross-talk even in a normal Ca2+/Mg2+-containing external medium. Presently it is un- known by which mechanism the astrocytic release of signaling mole- cules operates; exocytotic gliotransmitter release, outward passage through the dilated P2X7R itself or the associated protein Pax-1, as well as bestrophin1 channels are all possible exit pathways. Under patho- physiological conditions, such as ischemia, traumatic CNS injury and neurodegenerative illnesses (e.g. Parkinson’s Disease, Alzheimer’s Disease, amyotrophic lateral sclerosis, etc.) microglial P2X7Rs become activated by the large ATP concentrations leaving the cells damaged by the primary disease (Burnstock et al., 2011; Sperlagh and Illes, 2014; Bhattacharya and Biber, 2016). Then, microglia secretes various ex- cytotoXic/injurious mediators (e.g. glutamate, ATP, reactive oXygen and nitrogen species, etc.) onto neighboring neurons and causes mas- sive necrosis/cell death.

References

Abbracchio, M.P., Burnstock, G., 1994. Purinoceptors: are there A-438079 families of P2X and P2Y purinoceptors? Pharmacol. Ther. 64 (3), 445–475.
Alberto, A.V., Faria, R.X., Couto, C.G., Ferreira, L.G., Souza, C.A., TeiXeira, P.C., Froes, M.M., Alves, L.A., 2013. Is pannexin the pore associated with the P2X7 receptor? Naunyn Schmiedebergs Arch. Pharmacol. 386 (9), 775–787.
Armstrong, J.N., Brust, T.B., Lewis, R.G., MacVicar, B.A., 2002. Activation of presynaptic P2X7-like receptors depresses mossy ﬁber-CA3 synaptic transmission through p38 mitogen-activated protein kinase. J. Neurosci. 22 (14), 5938–5945.
Avishai-Eliner, S., Brunson, K.L., Sandman, C.A., Baram, T.Z., 2002. Stressed-out, or in (utero)? Trends Neurosci. 25 (10), 518–524.
Beamer, E., Goloncser, F., Horvath, G., Beko, K., Otrokocsi, L., Kovanyi, B., Sperlagh, B., 2016. Purinergic mechanisms in neuroinﬂammation: an update from molecules to behavior. Neuropharmacology 104, 94–104.
Bhattacharya, A., Biber, K., 2016. The microglial ATP-gated ion channel P2X7 as a CNS drug target. Glia 64 (10), 1772–1787.
Boucsein, C., Kettenmann, H., Nolte, C., 2000. Electrophysiological properties of micro-glial cells in normal and pathologic rat brain slices. Eur. J. Neurosci. 12 (6), 2049–2058.
Bowser, D.N., Khakh, B.S., 2007. Two forms of single-vesicle astrocyte exocytosis imaged with total internal reﬂection ﬂuorescence microscopy. Proc. Natl. Acad. Sci. U. S. A. 104 (10), 4212–4217.
Brockhaus, J., Ilschner, S., Banati, R.B., Kettenmann, H., 1993. Membrane properties of ameboid microglial cells in the corpus callosum slice from early postnatal mice. J. Neurosci. 13 (10), 4412–4421.
Burnstock, G., 2006. Historical review: ATP as a neurotransmitter. Trends Pharmacol. Sci. 27 (3), 166–176.
Burnstock, G., Verkhratsky, A., 2009. Evolutionary origins of the purinergic signalling system. Acta Physiol. (OXf.) 195 (4), 415–447.
Burnstock, G., Krugel, U., Abbracchio, M.P., Illes, P., 2011. Purinergic signalling: from normal behaviour to pathological brain function. Prog. Neurobiol. 95 (2), 229–274.
Chessell, I.P., Hatcher, J.P., Bountra, C., Michel, A.D., Hughes, J.P., Green, P., Egerton, J., Murﬁn, M., Richardson, J., Peck, W.L., Grahames, C.B., Casula, M.A., Yiangou, Y., Birch, R., Anand, P., Buell, G.N., 2005. Disruption of the P2X7 purinoceptor gene abolishes chronic inﬂammatory and neuropathic pain. Pain 114 (3), 386–396.
Cheung, G., Chever, O., Rouach, N., 2014. Connexons and pannexons: newcomers in neurophysiology. Front. Cell. Neurosci. 8, 348.
Clarke, D.D., 1991. Fluoroacetate and ﬂuorocitrate: mechanism of action. Neurochem. Res. 16 (9), 1055–1058.
Coco, S., Calegari, F., Pravettoni, E., Pozzi, D., Taverna, E., Rosa, P., Matteoli, M., Verderio, C., 2003. Storage and release of ATP from astrocytes in culture. J. Biol. Chem. 278 (2), 1354–1362.
Coddou, C., Yan, Z., Obsil, T., Huidobro-Toro, J.P., Stojilkovic, S.S., 2011. Activation and regulation of purinergic P2X receptor channels. Pharmacol. Rev. 63 (3), 641–683. Dahl, G., 2015. ATP release through pannexon channels. Philos. Trans. R. Soc. Lond. B Biol. Sci. 370 (1672).
Duan, S., Anderson, C.M., Keung, E.C., Chen, Y., Chen, Y., Swanson, R.A., 2003. P2X7 receptor-mediated release of excitatory amino acids from astrocytes. J. Neurosci. 23 (4), 1320–1328.
Ekstrom, A.D., Ranganath, C., 2017. Space, time, and episodic memory: the hippocampus is all over the cognitive map. Hippocampus (in press).
Fellin, T., Pozzan, T., Carmignoto, G., 2006. Purinergic receptors mediate two distinct glutamate release pathways in hippocampal astrocytes. J. Biol. Chem. 281 (7), 4274–4284.
Ficker, C., Rozmer, K., Kato, E., Ando, R.D., Schumann, L., Krugel, U., Franke, H., Sperlagh, B., Riedel, T., Illes, P., 2014. Astrocyte-neuron interaction in the substantia gelatinosa of the spinal cord dorsal horn via P2X7 receptor-mediated release of glutamate and reactive oXygen species. Glia 62 (10), 1671–1686.
Franke, H., Illes, P., 2014. Nucleotide signaling in astrogliosis. Neurosci. Lett. 565, 14–22.
Fredholm, B.B., Abbracchio, M.P., Burnstock, G., Daly, J.W., Harden, T.K., Jacobson, K.A., Leﬀ, P., Williams, M., 1994. Nomenclature and classiﬁcation of purinoceptors. Pharmacol. Rev. 46 (2), 143–156.
Freund, T.F., Buzsáki, G., 1996. Interneurons of the hippocampus. Hippocampus 6 (4), 347–470.
Fujita, T., Chen, M.J., Li, B., Smith, N.A., Peng, W., Sun, W., Toner, M.J., Kress, B.T., Wang, L., Benraiss, A., Takano, T., Wang, S., Nedergaard, M., 2014. Neuronal transgene expression in dominant-negative SNARE mice. J. Neurosci. 34 (50), 16594–16604.
Giaume, C., Leybaert, L., Naus, C.C., Saez, J.C., 2013. Connexin and pannexin hemi-channels in brain glial cells: properties, pharmacology, and roles. Front. Pharmacol. 4, 88.
Huidobro-Toro, J.P., Lorca, R.A., Coddou, C., 2008. Trace metals in the brain: allosteric modulators of ligand-gated receptor channels, the case of ATP-gated P2X receptors. Eur. Biophys. J. 37 (3), 301–314.
Illes, P., Alexandre Ribeiro, J., 2004. Molecular physiology of P2 receptors in the central nervous system. Eur. J. Pharmacol. 483 (1), 5–17.
Illes, P., Verkhratsky, A., 2016. Purinergic neurone-glia signalling in cognitive-related pathologies. Neuropharmacology 104, 62–75.
Illes, P., Verkhratsky, A., Burnstock, G., Franke, H., 2012. P2X receptors and their roles in astroglia in the central and peripheral nervous system. Neuroscientist 18 (5), 422–438.
Illes, P., Khan, T.M., Rubini, P., 2017. Neuronal P2X7 receptors revisited: do they really exist? J. Neurosci. 37 (30), 7049–7062.
Imura, Y., Morizawa, Y., Komatsu, R., Shibata, K., Shinozaki, Y., Kasai, H., Moriishi, K., Moriyama, Y., Koizumi, S., 2013. Microglia release ATP by exocytosis. Glia 61 (8), 1320–1330.
Jarvis, M.F., Khakh, B.S., 2009. ATP-gated P2X cation-channels. Neuropharmacology 56 (1), 208–215.
Jimenez-Pacheco, A., Diaz-Hernandez, M., Arribas-Blazquez, M., Sanz-Rodriguez, A., Olivos-Ore, L.A., Artalejo, A.R., Alves, M., Letavic, M., Miras-Portugal, M.T., Conroy, R.M., Delanty, N., Farrell, M.A., O’Brien, D.F., Bhattacharya, A., Engel, T., Henshall, D.C., 2016. Transient P2X7 receptor antagonism produces lasting reductions in spontaneous seizures and gliosis in experimental temporal lobe epilepsy. J. Neurosci. 36 (22), 5920–5932.
Jo, S., Yarishkin, O., Hwang, Y.J., Chun, Y.E., Park, M., Woo, D.H., Bae, J.Y., Kim, T., Lee, J., Chun, H., Park, H., Leee, D.Y., Hong, J., Kim, H.Y., Oh, S.J., Park, S.J., Lee, H., Yoon, B.E., Kim, Y., Jeong, Y., Shim, I., et al., 2014. GABA from reactive astrocytes impairs memory in mouse models of Alzheimer’s disease. Nat. Med. 20 (8), 886–896.
Kang, J., Kang, N., Lovatt, D., Torres, A., Zhao, Z., Lin, J., Nedergaard, M., 2008. Connexin 43 hemichannels are permeable to ATP. J. Neurosci. 28 (18), 4702–4711.
Khakh, B.S., 2009. ATP-gated P2X receptors on excitatory nerve terminals onto inter- neurons initiate a form of asynchronous glutamate release. Neuropharmacology 56 (1), 216–222.
Khakh, B.S., Gittermann, D., Cockayne, D.A., Jones, A., 2003. ATP modulation of excitatory synapses onto interneurons. J. Neurosci. 23 (19), 7426–7437.
Boue-Grabot, E., Pankratov, Y., 2017. Modulation of central synapses by astrocyte-re- leased ATP and postsynaptic P2X receptors. Neural Plast. 2017, 9454275.
Kukley, M., Stausberg, P., Adelmann, G., Chessell, I.P., Dietrich, D., 2004. Ecto-nucleo- tidases and nucleoside transporters mediate activation of adenosine receptors on hippocampal mossy ﬁbers by P2X7 receptor agonist 2’-3’-O-(4-benzoylbenzoyl)-ATP. J. Neurosci. 24 (32), 7128–7139.
Lalo, U., Palygin, O., Rasooli-Nejad, S., Andrew, J., Haydon, P.G., Pankratov, Y., 2014.
EXocytosis of ATP from astrocytes modulates phasic and tonic inhibition in the neocortex. PLoS Biol. 12 (1), e1001747.
Leybaert, L., Braet, K., Vandamme, W., Cabooter, L., Martin, P.E., Evans, W.H., 2003. Connexin channels, connexin mimetic peptides and ATP release. Cell Commun. Adhes. 10 (4-6), 251–257.
Lomo, T., 2018. Discovering long-term potentiation (LTP) – recollections and reﬂections on what came after. Acta Physiol. (OXf.) 222 (2).
Marques-da-Silva, C., Chaves, M.M., Castro, N.G., Coutinho-Silva, R., Guimaraes, M.Z., 2011. Colchicine inhibits cationic dye uptake induced by ATP in P2X2 and P2X7 receptor-expressing cells: implications for its therapeutic action. Br. J. Pharmacol. 163 (5), 912–926.
Masin, M., Young, C., Lim, K., Barnes, S.J., Xu, X.J., Marschall, V., Brutkowski, W., Mooney, E.R., Gorecki, D.C., Murrell-Lagnado, R., 2012. EXpression, assembly and function of novel C-terminal truncated variants of the mouse P2X7 receptor: re- evaluation of P2X7 knockouts. Br. J. Pharmacol. 165 (4), 978–993.
Metzger, M.W., Walser, S.M., Aprile-Garcia, F., Dedic, N., Chen, A., Holsboer, F., Arzt, E., Wurst, W., Deussing, J.M., 2017a. Genetically dissecting P2rX7 expression within the central nervous system using conditional humanized mice. Purinergic Signal. 13 (2), 153–170.
Metzger, M.W., Walser, S.M., Dedic, N., Aprile-Garcia, F., Jakubcakova, V., Adamczyk, M., Webb, K.J., Uhr, M., Refojo, D., Schmidt, M.V., Friess, E., Steiger, A., Kimura, M., Chen, A., Holsboer, F., Arzt, E., Wurst, W., Deussing, J.M., 2017b. Heterozygosity for the mood disorder-associated variant Gln460Arg alters P2X7 receptor function and sleep quality. J. Neurosci. 37 (48), 11688–11700.
Miras-Portugal, M.T., Gomez-Villafuertes, R., GualiX, J., Diaz-Hernandez, J.I., Artalejo, A.R., Ortega, F., Delicado, E.G., Perez-Sen, R., 2016. Nucleotides in neuroregenera- tion and neuroprotection. Neuropharmacology 104, 243–254.
Miras-Portugal, M.T., Sebastian-Serrano, A., de Diego, G.L., Diaz-Hernandez, M., 2017.
Neuronal P2X7 receptor: involvement in neuronal physiology and pathology. J. Neurosci. 37 (30), 7063–7072.
Montero, T.D., Orellana, J.A., 2015. Hemichannels: new pathways for gliotransmitter release. Neuroscience 286, 45–59.
Nedergaard, M., Verkhratsky, A., 2012. Artifact versus reality—how astrocytes contribute to synaptic events. Glia 60 (7), 1013–1023.
Neves, G., Cooke, S.F., Bliss, T.V., 2008. Synaptic plasticity, memory and the hippo- campus: a neural network approach to causality. Nat. Rev. Neurosci. 9 (1), 65–75.
Nicke, A., Kuan, Y.H., Masin, M., Rettinger, J., Marquez-Klaka, B., Bender, O., Gorecki, D.C., Murrell-Lagnado, R.D., Soto, F., 2009. A functional P2X7 splice variant with an alternative transmembrane domain 1 escapes gene inactivation in P2X7 knock-out mice. J. Biol. Chem. 284 (38), 25813–25822.
Nicoll, R.A., 2017. A brief history of long-term potentiation. Neuron 93 (2), 281–290.
Norenberg, W., Schunk, J., Fischer, W., Sobottka, H., Riedel, T., Oliveira, J.F., Franke, H., Illes, P., 2010. Electrophysiological classiﬁcation of P2X7 receptors in rat cultured neocortical astroglia. Br. J. Pharmacol. 160 (8), 1941–1952.
North, R.A., 2002. Molecular physiology of P2X receptors. Physiol. Rev. 82 (4), 1013–1067.
Oliveira, J.F., Riedel, T., Leichsenring, A., Heine, C., Franke, H., Krugel, U., Norenberg, W., Illes, P., 2011. Rodent cortical astroglia express in situ functional P2X7 receptors sensing pathologically high ATP concentrations. Cereb. Cortex 21 (4), 806–820.
Pankratov, Y., Lalo, U., 2014. Calcium permeability of ligand-gated Ca2+ channels. Eur. J. Pharmacol. 739, 60–73.
Pankratov, Y., Castro, E., Miras-Portugal, M.T., Krishtal, O., 1998. A purinergic compo- nent of the excitatory postsynaptic current mediated by P2X receptors in the CA1 neurons of the rat hippocampus. Eur. J. Neurosci. 10 (12), 3898–3902.
Pankratov, Y.V., Lalo, U.V., Krishtal, O.A., 2002. Role for P2X receptors in long-term potentiation. J. Neurosci. 22 (19), 8363–8369.
Parpura, V., Fang, Y., Basarsky, T., Jahn, R., Haydon, P.G., 1995. EXpression of sy- naptobrevin II, cellubrevin and syntaxin but not SNAP-25 in cultured astrocytes. FEBS Lett. 377 (3), 489–492.
Pelegrin, P., Surprenant, A., 2006. Pannexin-1 mediates large pore formation and inter- leukin-1β release by the ATP-gated P2X7 receptor. EMBO J. 25 (21), 5071–5082.
Rasooli-Nejad, S., Palygin, O., Lalo, U., Pankratov, Y., 2014. Cannabinoid receptors contribute to astroglial Ca2+-signalling and control of synaptic plasticity in the neocortex. Philos. Trans. R. Soc. Lond. B Biol. Sci. 369 (1654), 20140077.
Solle, M., Labasi, J., PerregauX, D.G., Stam, E., Petrushova, N., Koller, B.H., Griﬃths, R.J., Gabel, C.A., 2001. Altered cytokine production in mice lacking P2X7 receptors. J. Biol. Chem. 276 (1), 125–132.
Sperlagh, B., Illes, P., 2014. P2X7 receptor: an emerging target in central nervous system diseases. Trends Pharmacol. Sci. 35 (10), 537–547.
Suadicani, S.O., Brosnan, C.F., Scemes, E., 2006. P2X7 receptors mediate ATP release and ampliﬁcation of astrocytic intercellular Ca2+ signaling. J. Neurosci. 26 (5), 1378–1385.
Suadicani, S.O., Iglesias, R., Spray, D.C., Scemes, E., 2009. Point mutation in the mouse P2X7 receptor aﬀects intercellular calcium waves in astrocytes. ASN Neuro 1 (1). Surprenant, A., Rassendren, F., Kawashima, E., North, R.A., Buell, G., 1996. The cytolytic P2Z receptor for extracellular ATP identiﬁed as a P2X receptor (P2X7). Science 272 (5262), 735–738.
Virginio, C., Church, D., North, R.A., Surprenant, A., 1997. Eﬀects of divalent cations, protons and calmidazolium at the rat P2X7 receptor. Neuropharmacology 36 (9), 1285–1294.
Volonte, C., Apolloni, S., Skaper, S.D., Burnstock, G., 2012. P2X7 receptors: channels, pores and more. CNS Neurol. Disord. Drug Targets 11 (6), 705–721.
Voss, J.L., Bridge, D.J., Cohen, N.J., Walker, J.A., 2017. A closer look at the hippocampus and memory. Trends Cogn. Sci. 21 (8), 577–588.
Woo, J., Min, J.O., Kang, D.S., Kim, Y.S., Jung, G.H., Park, H.J., Kim, S., An, H., Kwon, J., Kim, J., Shim, I., Kim, H.G., Lee, C.J., Yoon, B.E., 2018. Control of motor coordination by astrocytic tonic GABA release through modulation of excitation/inhibition balance in cerebellum. Proc. Natl. Acad. Sci. U. S. A. 115 (19), 5004–5009.
Yan, Z., Khadra, A., Li, S., Tomic, M., Sherman, A., Stojilkovic, S.S., 2010. Experimental characterization and mathematical modeling of P2X7 receptor channel gating. J. Neurosci. 30 (42), 14213–14224.
Zhang, Q., Haydon, P.G., 2005. Roles for gliotransmission in the nervous system. J. Neural Transm. (Vienna) 112 (1), 121–125.